Regression

Which countries are most likely to fight wars?

This study is a reproduction of the following article published in the Economist:

Article: The Economist Nov 8th 2018 edition - Which countries are most likely to fight wars?

The code and data is available here:

Data Source: The Economist GitHub

First, let’s take a look at Combat deaths by region over the past century:

Creating logistic regressions to estimate the probability that a country would be involved in a conflict in any year, given its level of wealth.

The third-order polynomial equations is selected to predict the probability of fighting a war in a given year

country_involved_in_conflict ~ wealth + wealth^2 + wealth^3).

## 
## Call:
## glm(formula = country_involved_in_conflict ~ log2_interpolated_gdp_per_capita + 
##     log2_interpolated_gdp_per_capita_squared + log2_interpolated_gdp_per_capita_cubed, 
##     family = "binomial", data = df)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -0.5177  -0.5000  -0.4613  -0.3996   2.8354  
## 
## Coefficients:
##                                           Estimate Std. Error z value Pr(>|z|)
## (Intercept)                               4.216077   8.333871   0.506   0.6129
## log2_interpolated_gdp_per_capita         -2.823157   2.136572  -1.321   0.1864
## log2_interpolated_gdp_per_capita_squared  0.342850   0.180959   1.895   0.0581
## log2_interpolated_gdp_per_capita_cubed   -0.012530   0.005063  -2.475   0.0133
##                                           
## (Intercept)                               
## log2_interpolated_gdp_per_capita          
## log2_interpolated_gdp_per_capita_squared .
## log2_interpolated_gdp_per_capita_cubed   *
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 11015  on 17094  degrees of freedom
## Residual deviance: 10852  on 17091  degrees of freedom
## AIC: 10860
## 
## Number of Fisher Scoring iterations: 6

Regression

Which countries are most likely to fight wars?

This study is a reproduction of the following article published in the Economist:

The code and data is available here:

First, let’s take a look at Combat deaths by region over the past century:

Creating logistic regressions to estimate the probability that a country would be involved in a conflict in any year, given its level of wealth.

Wealth vs chance of conflict

References

The citations and data sources used for this case